ABSTRACT
It has been half a century since the launch of the ¦rst-generation Earth observation satellites, such as former Soviet Union’s Sputnik 1 and the United States’ Television Infrared Observation Satellite Program (TIROS)-1 (Tatem et al., 2008). Compared with
traditional observation approaches, one important advantage of satellite remote sensing is its global mapping capability. Satellite data have thus become an unprecedented source to study global environmental changes (e.g., Kaufman et al., 2002; Nemani et al., 2003). Ÿanks to the continuity of satellite missions, the length of remote sensing data record is continuously increasing. Currently, we have 30+ years archive of the National Oceanic and Atmospheric Administration (NOAA) advanced very-highresolution radiometer (AVHRR) at spatial resolution suitable for continental or global studies (Beck et al., 2011) and Landsat TM/ ETM+ data at ecosystem scales (Markham and Helder, 2012). Entering the new century, data records from new-generation remote sensors such as the Satellite Pour l’Observation de la Terre (SPOT) VEGETATION are also lengthening. More than one decade of Moderate Resolution Imaging Spectroradiometer (MODIS) data become available (2000-present) and serve as valuable resource for the atmospheric (King et al., 2003), terrestrial (Justice et al., 2002), and oceanic (Esaias et al., 1998) research community with well-designed spectral con¦guration and improved radiometric and geometric accuracy (Justice et al., 1998).
